What does 0 dB mean?

  • Thread starter Thread starter skim
  • Start date Start date
S

skim

New member
Can someone explain exactly what it means when my audio clips (the red light turns on)? Why is it that when it "clips", I don't hear it "clipping" on my monitors or headphones? What will it actually mean when audio samples that clip are transferred to a CD and played in a stereo?

I'm asking this because I'm trying to increase levels on a recording using a limiter. I've tried to get it as aggressive as possible, and my track is hitting 0 on numerous occasions. But when I put it on a CD and play it in my car, I still have to turn the volume up twice as loud as a commercial production CD.

Thanks -

Steven
 
0dBfs - The fs is Full Scale - Can't go any higher.

The reason it's not as loud is that there's a difference between smashing with a limiter and a proper mastering session. ALTHOUGH the mix, the recording, even the raw sound and playing style will in the end, dictate the "maximum listening level" of a recording.

That level is determined long before the RECORD button is ever pressed.
 
Don't mean to hijack this thread, but Massive can I ask you to expand your answer. What I am guessing is, you can have clipping caused by a narrow frequency range, but the main "body" is well below clipping. A true ME would put more of the full freq spectrum right up to clip, thus making the music "louder" (and fuller). Is that close?

Thanks.
 
Okay, that second answer at least made some sense to me. Can I once post a newbie question and not get bashed with a "if you're not a pro then fugeddit" answer? Yes, you're a pro and you deserve your money, let go of the insecurity already. I was asking for an explanation. If you think I'm too dumb to understand your best efforts at dummying this down for me, then don't bother replying.

So how would a newbie like me with no money go about trying to raise the levels of a mix? Is moderate clipping then considered acceptable?

I've heard what I think clipping is. It distorts. How come I didn't get that distortion sound on my CD with the tracks that supposedly hit 0 dB?

Thank you again.
 
Skim, I'm not saying to fuggedaboutit - If you want to experiment with loudness techniques, please go ahead. I'm just stating that pushing the input into a limiter is normally not going to make things much better - It hurts more than it helps.

I apologize if my comment offended - Not what I was intending. I know that I seem to appear a bit "short-fused" on this topic. This is a very sensitive subject for me. A lot of otherwise great sounding recordings are being hammered into the ground for the sake of sheer volume and it truly sickens me.

Anyway, the point is that there is no magic bullet or technique - You're trying to stack nickels to make dollars. Every 1/2dB EQ adjustment - Mid-side, parallel compression, phase adjustment, harmonic saturation... They ALL are crucial to the overall volume of the finished product. All have to be taken into account, and all have to be in harmony with everything else. Pushing a limiter adds volume until it adds distortion. Then, it only adds distortion.

If absolutely nothing else, put the limiter at the end of your chain (before dithering, of course) and set it to -0.5dB. Then forget that it's there. All it is is a brick wall that nothing should penetrate (although things can and will, but that's for another thread).

Use a compressor - Maybe even (God forgive me) a multi-band compressor if you have one. Experiment with every control on it. Use THAT to get some extra volume.

HOWEVER - As stated before, the maximum apparent level you can attain is STILL determined before the recording starts. There are plenty of recordings I get in here that will NEVER be as loud as a commercial CD. They don't have the potential, period. Maybe they used the wrong guitar tone. Maybe the drum mix is too loose. Maybe the vocals are too loud. Again, stacking nickels to make dollars. Everything has to be right BEFORE and DURING the session for it to be loud afterwards.
 
Hey I appreciate the reply. Actually when I wrote the first question I was going to write a line like "I know everyone's sensitive about compressors here, but..." then decided not to waste space. Also after I wrote my second post I felt a bit short-fused too, so I apologize.

Funny, you guys are tired of compressor newbies, and I'm tired of you guys being tired of compressor newbies. I'm thinking now that I should just get some estimates from the studios I work with to get this mastered. I do recognize the art that is audio engineering, just sometimes don't have the money (and personally want to learn too).

Steven
 
Learning is a good thing.

Learning mastering technique is almost always easier and better on other people's material - I (and many other M.E.s) have a "foot tapping" point - Once you start tapping your foot to it, you can't work objectively anymore. Familiarity is bad.

Mastering your own recordings (I'm gonna catch hell for this - just wait) "normally" isn't the best way to go - ESPECIALLY on the same monitors it was mixed on. Long story on that (if you're not busy, download the Guide to MIXING for MASTERING from the MM site).

Whenever I have to master something I've mixed, well, that's why I keep several sets of monitors in-house. Every monitor / space / engineer has deficiencies somewhere - Mastering in the same space by the same person on the same monitors that something was mixed on multiplies the PROBLEMS. If you're not a big fan of 400Hz, you cut it during recording. Then more during mixing. Then, you compress during mastering and bring it up again, so you cut it again. The same thing goes for room nodes and speaker deficiencies. The more familiar you are with the material, the more bold you will be with adjustments. Monitor-wise, I'll work backwards - When I'm mastering, I switch over to studio monitors every now and then just to see how everything translates. When I'm mixing, I'm in the studio monitors. Every now and then, I'll switch over to my mastering system to see how things translate.

If you can, get a friend or other engineer with ears that you trust in the room with you - I bring in another engineer or at least my apprentice any time I'm working on material that I'm familiar with just as a "touch of reality." I've/they've caught me making some unusual adjustments before, and I'm sure they will again.
 
Yo Skim! aside from what Massive Master said, which is the voice of experience talking, the "clip" light on most preamps is a warning, not a statement that clipping is occuring. Most preamps have some "headroom", an amount beyond which the pre can be pushed without clipping after that light comes on. Typically, that's about 6db, but it can be more or less. That's one of the features of better preamps, they often have more headroom.
My Avalon AD2022, for instance, is wicked hard to clip. When the level meters are pegged, and that red light comes on, I've still got 32db of headroom! This means you pretty much have to ignore all reason to clip the thing. My TC Electronics M300 (an FX box), on the other hand, has about 3db. When that red light comes on, you're already in trouble. Just remember, the red light isn't telling you you are clipping, but that you are about to. You are probably staying within the margin of safety provided by the headroom of the unit you are using. If you have good ears, when you really clip that puppy, you'll know it.-Richie
 
Unlike the analog domain, there is a hard limit to the voltage your analog to digital converters can translate to numbers. If you think of a sine wave, when the amplitude gets too high for your soundcard's converters, the top gets flattened out, or even worse, starts showing up at the bottom. The frequency content of that section where the peak is chopped off is full of nasty harmonics, and it sounds poopy. In analog equipment, especially tubes and tape, when you start exceeding the linear limit of the equipment, it rounds off the top of the peak, and those harmonics are usually pleasant. If you have overloaded digitally for a millisecond, you may not be able to hear it. But it is still very bad, and should be avoided.
 
its nice to see people apologize and learn versus suing someone for $5k plus for "emotional distress" oh yeah and when i came in here...would you believe i never used a compressor during recording or after? i use one now but not on the final mix or my attempt at mastering
 
well I learned something from this discussion, and that is that I don't know anything about this subject. Where is the best place for MM info, is there a faq? books?

It clears up the red light thing and the meters thing. Manuals usually say that you don't want the clipping lights or the 'red area' or whatever meter they use to be on a lot. That always leaves a lot of interpretation and I always wondered how much of the time I could let those lights, or top meter area stay up or on. So it's just a warning and I use my ears - it's different on my different equipment b/c of the difference in calibration for one thing I suppose, but also b/c of different amounts of headroom. I did notice that on my little MR8 and older MD format digital stuff that I can push it up more and not distort and have higher level in the final mix.

This is all very interesting stuff and seems to be a big part of separating the 'home' recordings from the pro, those techniques that are under the surface and not in the mainstream advertising for the 'home' gear. All the 'home' stuff ads seem to suggest that one can create a final product all in the same unit. The 'pros' probably get 'sickened' when they read those ads....
 
In a simple sense I would think of clipping this way. Clipping a signal robs the audio completely of any dynamic range. It essentially chops off everything that has reached the 0Dbfs limit.

It is better to apply compression first so that transients are reduced by a given amount without killing off all of the transient material. Using soft-knee also helps to preserve the range a bit more by making the compression react in a smoother manner than just hitting it a prescribed level.

Compression by it's nature does not add level however, it reduces it. It's the gain makeup that allows the average volume of the material to increase. Since you've reduced the peaks by a given amount, you can now raise the overall level by the amount that you have left after compression up to the 0Dbfs limit.

Even with compression, there can be ocassional peaks that are outside of the "average". This is where a limiter is useful. There may be stray snare or kick transients that are robbing your ability to raise the overall volume by a few Db, the ear is less sensitive to crushing these shorter transients, so chopping them off is a good compromise if you want to raise the overall volume.

The issue that many MEs have is that this type of limiting has been abused, and is killing off more transient material than it should for that sake of making a CD loud. Since most folks are going to lower the volume anyway, it's kinda stupid because the hypercompressed audio will lack the quality that a less compressed CD will have if played at the same volume.

Here's a simple recipe, first use a compressor to add density to the sound. Play around with the threshold, ratio, attack, release, and any soft-limit feature you may have available. Forget about volume at this point. Just get it to sound as "thick" as you want. Next use make-up gain to get the entire mix close to the 0Dbfs limit. If you still feel it needs volume, than try adding a limiter "rinse and repeat".

In addition to dynamics, EQ plays an important role to all of this. But that's another thread altogether ...
 
Thanks for the info and links. BTW, I assumed my tracks were also lower levels than commercial CD's but I've been listening to a few comparisons for the past two hours or so and actually, my CD's aren't any lower levels I've kept the volume constant, of course. And, on some of my solo acoustic w/ vocal, I'm even louder than some, and w/ out much perceptible distortion thru a home stereo system.

I also noticed on some commerical CD's that they sound smaller. I guess compressed lke crazy and also very limited in the freq.s of each instrument. Whereas I can keep my vocal and my acoustic guitar fat with little eq bandwidth rolling off on either end, or no eq. I guess I don't need to do that w/ few interfering freq. ranges of the other missing instruments?

On my older rec.s the acoustic is boomy and I hate that. Later on I've placed the mic over the guit. a foot above, in the plane of the top above 12th fret or so and also switched from dynamic Sure to a MXL v67 w/ Art preamp and I'm not boomy anymore.
 
You must log in or register to reply here.
Back
Top